賴美淑臺灣大學：預防醫學研究所謝孟書Hsieh, Meng-ShuMeng-ShuHsieh2007-11-282018-06-292007-11-282018-06-292004http://ntur.lib.ntu.edu.tw//handle/246246/59230研究背景：糖尿病與原發性肝癌的相關性在西方國家已經被證實，但是兩者之間的關係在肝炎高盛行區仍需被檢驗，糖尿病與其他危險因子之間可能的效果修飾作用或交互作用也需要作進一步的探討。 方法：本研究為前瞻性世代研究，來自基隆社區複合式篩檢的54929位受檢者自1999年追蹤至2003年，肝癌病例由二階段肝癌篩檢及癌症登記確定，危險因子則由抽血及問卷得到。 結果：基隆社區30歲以上5萬多人前瞻世代,複合式篩檢,於1999 ~ 2003共發現138位肝癌的病例，在控制了年齡、性別、B型肝炎、C型肝炎等重要危險因子後，第二型糖尿病對肝癌的影響沒有統計上的顯著。對此影響，C型肝炎(P=0.04)及高膽固醇血(P=0.03)兩個有顯著的效果修飾作用，但是協同指數(synergy index)只有糖尿病與高膽固醇血的粗勝算比(crude odds ratio)才有顯著性。分層分析發現糖尿病顯著地增加肝癌危險性的情況為沒有C型肝炎者(adjusted odds ratio=1.98)及高膽固醇血者(adjusted odds ratio=2.52)。 結論：在B型及C型肝炎盛行的台灣，糖尿病對肝癌的發生在控制了主要的危險因子後沒有顯著的影響，與之前在肝癌低盛行區的研究不同。此影響受到C型肝炎及高膽固醇血的修飾；在無C 型肝炎及高膽固醇血症病人群，糖尿病對肝癌的發生呈現有意義的相關。Background: The association between diabetes mellitus (DM) and primary hepatocellular carcinoma (HCC) in western countries had been corroborated. But the relation in countries with high prevalence of hepatitis should be examined. The possible effect modification or interaction between DM and other risk factors still need to be explored. Method: A prospective cohort with 54,929 subjects from the Keelung Community-based Integrated Screening (KCIS) was followed from 1999 to 2003. HCC was ascertained by two-stage liver screen and National Cancer Registry linkage. Risk factors were measured by blood sampling and questionnaire. Results: 138 HCC cases were identified. The effect of type 2 DM on HCC was not statistically significant after controlling age, gender, HBV and HCV that were strong risk factors. HCV (P=0.04) and hypercholesterolemia (P=0.03) were two significant effect modifiers. But the synergy index was significant only in crude odds ratio for interaction between DM and hypercholesterolemia. Stratifications which diabetes significantly increases the risk of HCC are patients without hepatitis C (adjusted odds ratio=1.98) and patients with hypercholesterolemia (adjusted odds ratio=2.52) Conclusion: Different form previous studies in the low HCC prevalence areas, diabetes mellitus does not play an independent role in the development of HCC in Taiwan after adjusting major risk factors. The association between DM and liver cancer was influenced or modified by hepatitis C and total cholesterol. The effect of DM for HCC is significant in the absence of hepatitis C infection and in the presence of hypercholesterolemia.I. Introduction 1 II. Literature review 4 1. Biological Mechanism on the Association between DM and HCC 4 (1) Insulin-like growth factor I (IGF-I) 4 (2) Insulin-like growth factor-binding protein 3(IGFBP-3) 5 (3) Lipid profile and other factors 5 2. Epidemiological Studies on Diabetes Mellitus and Liver Cancer 6 (1) Case-control study 8 (2) Cohort study 16 (3) Effect modification 19 III. Material and Methods 20 1. Study Design 20 2. Ascertainment of type 2 diabetes 22 3. Ascertainment of HCC 22 4. Risk Factors 24 5. Statistical Analysis 25 IV. Results 28 1. Annual Incidence of Hepatocelluar Carcinoma 28 2. Risk factors of HCC 28 3. Effect Modification 30 4. Association between Type 2 Diabetes and HCC by the Stratification of Hepatitis and Hypercholesterolemia 32 5. Cumulative Incidence Rate of HCC by Type 2 Diabetes and Its’ Effect Modifiers 33 V. Discussion & Conclusion 35 1. Discussion 35 2. Conclusion 40 Reference 42en-US肝癌交互作用世代研究第二型糖尿病B型肝炎C型肝炎cohort studyhepatocellular carcinomatype II diabeteshepatitis Chepatitis Binteraction[SDGs]SDG3thesis